diff --git a/Biobase/Primary.hs b/Biobase/Primary.hs --- a/Biobase/Primary.hs +++ b/Biobase/Primary.hs @@ -10,6 +10,7 @@ , module Biobase.Primary.Letter , module Biobase.Primary.Nuc , module Biobase.Primary.Trans + , module Biobase.Primary.Unknown ) where import Biobase.Primary.AA hiding (Stop,A,B,C,D,E,F,G,H,I,K,L,M,N,P,Q,R,S,T,V,W,X,Y,Z,Undef) @@ -18,4 +19,5 @@ import Biobase.Primary.Letter import Biobase.Primary.Nuc import Biobase.Primary.Trans +import Biobase.Primary.Unknown diff --git a/Biobase/Primary/AA.hs b/Biobase/Primary/AA.hs --- a/Biobase/Primary/AA.hs +++ b/Biobase/Primary/AA.hs @@ -128,25 +128,10 @@ succ (Letter x) = Letter $ x+1 pred Stop = error "pred/Stop:AA" pred (Letter x) = Letter $ x-1 - toEnum k | k>=0 && k<=(unLetter Undef) = Letter k + toEnum k | k>=0 && k<=(getLetter Undef) = Letter k toEnum k = error $ "toEnum/Letter RNA " ++ show k fromEnum (Letter k) = k -instance MkPrimary [Char] AA where - primary = VU.fromList . map charAA - -instance MkPrimary [Letter AA] AA where - primary = VU.fromList - instance MkPrimary (VU.Vector Char) AA where primary = VU.map charAA - -instance MkPrimary BS.ByteString AA where - primary = VU.fromList . map charAA . BS.unpack - -instance MkPrimary BSL.ByteString AA where - primary = VU.fromList . map charAA . BSL.unpack - -instance MkPrimary T.Text AA where - primary = VU.fromList . map charAA . T.unpack diff --git a/Biobase/Primary/Letter.hs b/Biobase/Primary/Letter.hs --- a/Biobase/Primary/Letter.hs +++ b/Biobase/Primary/Letter.hs @@ -5,6 +5,7 @@ module Biobase.Primary.Letter where +import Control.DeepSeq (NFData) import Data.Aeson import Data.Binary import Data.Hashable (Hashable) @@ -29,7 +30,7 @@ -- | A 'Letter' together with its phantom type @t@ encodes bio-sequences. -newtype Letter t = Letter { unLetter :: Int } +newtype Letter t = Letter { getLetter :: Int } deriving (Eq,Ord,Generic,Ix) instance Binary (Letter t) @@ -37,6 +38,8 @@ instance FromJSON (Letter t) instance ToJSON (Letter t) +instance NFData (Letter t) + type Primary t = VU.Vector (Letter t) -- | Conversion from a large number of sequence-like inputs to primary @@ -68,7 +71,7 @@ -- *** Instances for 'Letter'. derivingUnbox "Letter" - [t| forall a . Letter a -> Int |] [| unLetter |] [| Letter |] + [t| forall a . Letter a -> Int |] [| getLetter |] [| Letter |] instance Hashable (Letter t) @@ -109,50 +112,4 @@ {-# Inline streamUp #-} streamDown l h = map (\(Z:.k) -> k) $ streamDown (Z:.l) (Z:.h) {-# Inline streamDown #-} - -{- -instance (Index sh, Show sh) => Shape (sh :. Letter z) where - rank (sh:._) = rank sh + 1 - zeroDim = zeroDim:.Letter 0 - unitDim = unitDim:.Letter 1 -- TODO does this one make sense? - intersectDim (sh1:.n1) (sh2:.n2) = intersectDim sh1 sh2 :. min n1 n2 - addDim (sh1:.Letter n1) (sh2:.Letter n2) = addDim sh1 sh2 :. Letter (n1+n2) -- TODO will not necessarily yield a valid Letter - size (sh1:.Letter n) = size sh1 * n - sizeIsValid (sh1:.Letter n) = sizeIsValid (sh1:.n) - toIndex (sh1:.Letter sh2) (sh1':.Letter sh2') = toIndex (sh1:.sh2) (sh1':.sh2') - fromIndex (ds:.Letter d) n = fromIndex ds (n `quotInt` d) :. Letter r where - r | rank ds == 0 = n - | otherwise = n `remInt` d - inShapeRange (sh1:.n1) (sh2:.n2) (idx:.i) = i>=n1 && i error "empty list in shapeOfList/Primary" - x:xs -> shapeOfList xs :. Letter x - deepSeq (sh:.n) x = deepSeq sh (n `seq` x) - {-# INLINE rank #-} - {-# INLINE zeroDim #-} - {-# INLINE unitDim #-} - {-# INLINE intersectDim #-} - {-# INLINE addDim #-} - {-# INLINE size #-} - {-# INLINE sizeIsValid #-} - {-# INLINE toIndex #-} - {-# INLINE fromIndex #-} - {-# INLINE inShapeRange #-} - {-# INLINE listOfShape #-} - {-# INLINE shapeOfList #-} - {-# INLINE deepSeq #-} - -instance (Shape sh, Show sh, ExtShape sh) => ExtShape (sh :. Letter z) where - subDim (sh1:.Letter n1) (sh2:.Letter n2) = subDim sh1 sh2 :. Letter (n1-n2) - rangeList (sh1:.Letter n1) (sh2:.Letter n2) = [ sh:.Letter n | sh <- rangeList sh1 sh2, n <- [n1 .. (n1+n2)]] - rangeStream (fs:.Letter f) (ts:.Letter t) = VM.flatten mk step Unknown $ rangeStream fs ts where - mk sh = return (sh :. f) - step (sh :. k) - | k>t = return $ VM.Done - | otherwise = return $ VM.Yield (sh :. Letter k) (sh :. k +1) - {-# INLINE [1] mk #-} - {-# INLINE [1] step #-} - {-# INLINE rangeStream #-} --} diff --git a/Biobase/Primary/Unknown.hs b/Biobase/Primary/Unknown.hs new file mode 100644 --- /dev/null +++ b/Biobase/Primary/Unknown.hs @@ -0,0 +1,72 @@ + +-- | A 'Letter' with unknown annotation. We sometimes want to encode that +-- we are dealing with @Letter@s in an alphabet, but we do not want to +-- commit to a certain alphabet (just yet). +-- +-- This module allows us to make explicit that we do not know the specific +-- alphabet type yet. + +module Biobase.Primary.Unknown where + +import Control.Applicative ((<$>)) +import Control.Arrow ((***),first) +import Data.Hashable +import Data.Ix (Ix(..)) +import Data.Map.Strict (Map) +import Data.Primitive.Types +import Data.Tuple (swap) +import Data.Vector.Unboxed.Deriving +import Debug.Trace +import GHC.Base (remInt,quotInt) +import GHC.Generics (Generic) +import GHC.Read +import qualified Data.Bijection.Map as B +import qualified Data.ByteString.Char8 as BS +import qualified Data.ByteString.Lazy.Char8 as BSL +import qualified Data.Map.Strict as M +import qualified Data.Text as T +import qualified Data.Vector.Generic as VG +import qualified Data.Vector.Generic.Mutable as VGM +import qualified Data.Vector.Unboxed as VU +import qualified Text.ParserCombinators.ReadPrec as RP +import qualified Text.Read.Lex as Lex + +import Biobase.Primary.Letter + + + +-- | @Unknown@ phantom type. + +data Unknown + + + +-- | Creating an unknown letter. + +unk :: Int -> Letter Unknown +unk = Letter + + + +-- *** instances + +instance Show (Letter Unknown) where + show (Letter i) = "U " ++ show i + +instance Read (Letter Unknown) where + readPrec = parens $ do + Lex.Ident u <- lexP + case u of + "U" -> unk <$> readPrec + _ -> RP.pfail + +instance Enum (Letter Unknown) where + succ (Letter x) = Letter $ x+1 + pred (Letter x) = Letter $ x-1 + toEnum = Letter + fromEnum = getLetter + +instance MkPrimary (VU.Vector Int) Unknown where + primary = VU.map Letter + {-# Inline primary #-} + diff --git a/BiobaseXNA.cabal b/BiobaseXNA.cabal --- a/BiobaseXNA.cabal +++ b/BiobaseXNA.cabal @@ -1,8 +1,9 @@ name: BiobaseXNA -version: 0.9.1.1 +version: 0.9.2.0 author: Christian Hoener zu Siederdissen maintainer: choener@bioinf.uni-leipzig.de -homepage: http://www.bioinf.uni-leipzig.de/~choener/ +homepage: https://github.com/choener/BiobaseXNA +bug-reports: https://github.com/choener/BiobaseXNA/issues copyright: Christian Hoener zu Siederdissen, 2011 - 2015 category: Bioinformatics synopsis: Efficient RNA/DNA representations @@ -10,6 +11,7 @@ license-file: LICENSE build-type: Simple stability: experimental +tested-with: GHC == 7.8.4, GHC == 7.10.1 cabal-version: >= 1.10.0 description: This is a base library for bioinformatics with emphasis on RNA @@ -37,30 +39,33 @@ sources/iupac-nucleotides sources/codontable changelog.md + README.md -library - build-depends: base >= 4.7 && < 4.9 - , aeson == 0.8.* - , bimaps == 0.0.0.* - , binary == 0.7.* - , bytes == 0.15.* - , bytestring == 0.10.* - , cereal == 0.4.* - , cereal-vector == 0.2.* - , containers == 0.5.* - , csv == 0.1.* - , file-embed == 0.0.8.* - , hashable == 1.2.* - , lens == 4.* - , primitive >= 0.5 && < 0.7 - , PrimitiveArray == 0.6.* - , split == 0.2.* - , text == 1.* - , tuple == 0.3.* - , vector == 0.10.* - , vector-binary-instances == 0.2.* - , vector-th-unbox == 0.2.* + +library + build-depends: base >= 4.7 && < 4.9 + , aeson >= 0.8 && < 0.10 + , bimaps >= 0.0.0.2 && < 0.0.1.0 + , binary >= 0.7 && < 0.8 + , bytes >= 0.15 && < 0.16 + , bytestring >= 0.10 && < 0.11 + , cereal >= 0.4 && < 0.5 + , cereal-vector >= 0.2 && < 0.3 + , containers >= 0.5 && < 0.6 + , csv >= 0.1 && < 0.2 + , deepseq >= 1.3 && < 1.5 + , file-embed >= 0.0.8 && < 0.0.9 + , hashable >= 1.2 && < 1.3 + , lens >= 4.0 && < 4.13 + , primitive >= 0.5 && < 0.7 + , PrimitiveArray >= 0.6.0 && < 0.6.2 + , split >= 0.2 && < 0.3 + , text >= 1.0 && < 1.3 + , tuple >= 0.3 && < 0.4 + , vector >= 0.10 && < 0.11 + , vector-binary-instances >= 0.2 && < 0.3 + , vector-th-unbox >= 0.2 && < 0.3 exposed-modules: Biobase.Primary Biobase.Primary.AA @@ -74,6 +79,7 @@ Biobase.Primary.Nuc.RNA Biobase.Primary.Nuc.XNA Biobase.Primary.Trans + Biobase.Primary.Unknown Biobase.Secondary Biobase.Secondary.Basepair Biobase.Secondary.Constraint @@ -82,7 +88,6 @@ Biobase.Secondary.Pseudoknots Biobase.Secondary.Structure Biobase.Secondary.Vienna - default-extensions: BangPatterns , DeriveGeneric , EmptyDataDecls @@ -98,16 +103,17 @@ , TypeOperators , UndecidableInstances , ViewPatterns - - default-language: Haskell2010 - + default-language: + Haskell2010 ghc-options: -O2 -funbox-strict-fields + + executable SubOptDistance - build-depends: base - , cmdargs == 0.10.* - , BiobaseXNA + build-depends: base + , BiobaseXNA + , cmdargs >= 0.10 && < 0.11 main-is: SubOptDistance.hs hs-source-dirs: @@ -120,6 +126,8 @@ , ScopedTypeVariables ghc-options: -O2 + + source-repository head type: git diff --git a/README.md b/README.md new file mode 100644 --- /dev/null +++ b/README.md @@ -0,0 +1,15 @@ +[![Build Status](https://travis-ci.org/choener/BiobaseXNA.svg?branch=master)](https://travis-ci.org/choener/BiobaseXNA) + +# BiobaseXNA + +Efficient encoding of biological sequences. + + + +#### Contact + +Christian Hoener zu Siederdissen +Leipzig University, Leipzig, Germany +choener@bioinf.uni-leipzig.de +http://www.bioinf.uni-leipzig.de/~choener/ + diff --git a/changelog.md b/changelog.md --- a/changelog.md +++ b/changelog.md @@ -1,3 +1,8 @@ +0.9.2.0 +------- + +- fixed overlapping instances in AA + 0.9.1.0 -------